Inference

[lesecs]

I have this error in uvr v5 inference.py: error: argument --input/-i: expected one argument Dont know how to fix it.

ty

[brmarkus]

Which command-line have you called, of which application/example?

[lesecs]

Running MGM-v5-2Band-32000-BETA1 usage: inference.py [-h] [--gpu GPU] [--pretrained_model PRETRAINED_MODEL] --input INPUT [--nn_architecture NN_ARCHITECTURE] [--model_params MODEL_PARAMS] [--window_size WINDOW_SIZE] [--output_image] [--deepextraction] [--postprocess] [--tta] [--high_end_process {none,bypass,correlation}] [--aggressiveness AGGRESSIVENESS] inference.py: error: argument --input/-i: expected one argument

[brmarkus]

Can you show the full command-line, paths, parameters, please? Are you calling "python3 inference.py"... without parameters...?

[lesecs]

Run with TTA? [Y/N]: y

      Choose Your Model

=======================================

a. HP2-4BAND-3090_4band_1 b. HP2-4BAND-3090_4band_2 c. HP_4BAND_3090 d. Vocal_HP_4BAND_3090 e. Vocal_HP_4BAND_3090_AGG f. Ensemble All HP Instrumental Models (5 Models) - Final Outputs Save to Ensembled Folder! g. Ensemble HP2 Models Only (2 Models) - Final Outputs Save to Ensembled Folder! Type the letter associated with the option you choose and hit 'Enter': a Running HP2-4BAND-3090_4band_1 usage: inference.py [-h] [--gpu GPU] [--pretrained_model PRETRAINED_MODEL] --input INPUT [--nn_architecture NN_ARCHITECTURE] [--model_params MODEL_PARAMS] [--window_size WINDOW_SIZE] [--output_image] [--deepextraction] [--postprocess] [--tta] [--high_end_process {none,bypass,correlation}] [--aggressiveness AGGRESSIVENESS] inference.py: error: argument --input/-i: expected one argument

[lesecs]

can you tell me where i can find and download the last uvr v5 beta. May be i dont have the last one. ty

[lesecs]

I use python 3.7.9 amd64

[lesecs]

Run with TTA? [Y/N]: y

      Choose Your Model

=======================================

a. HP2-4BAND-3090_4band_1 b. HP2-4BAND-3090_4band_2 c. HP_4BAND_3090 d. Vocal_HP_4BAND_3090 e. Vocal_HP_4BAND_3090_AGG f. Ensemble All HP Instrumental Models (5 Models) - Final Outputs Save to Ensembled Folder! g. Ensemble HP2 Models Only (2 Models) - Final Outputs Save to Ensembled Folder! Type the letter associated with the option you choose and hit 'Enter': a Running HP2-4BAND-3090_4band_1 loading model... Traceback (most recent call last): File "inference.py", line 273, in main() File "inference.py", line 133, in main model = nets.CascadedASPPNet(mp.param['bins'] * 2) UnboundLocalError: local variable 'nets' referenced before assignment